There is significant comorbidity between alcohol, drug and mental disorders in treatment samples and the general population. The basis of this phenomenon is not clear. The co-occurrence of these disorders could be due to a common underlying genetic vulnerability or it could be the result of one disorder serving as an environmental influence that predisposes an individual to the other disorders. The elucidation of the etiology of the comorbidity would be important for prevention. Twin cross-concordant data provide a method for assessing the cause of observed comorbidity. Twins are cross-concordant if one twin is affected with one disorder and the other twin is affected with a second disorder. If there are genetic factors common to both disorders, identical twins would be cross-concordant more often than would fraternal twins. If a significant difference is not found between identical and fraternal twins, the observed relationship between disorders is likely due to environmental factors. Interpreting cross-concordance data is made difficult by two factors. First, cross-concordance may reflect the independent concordance for the second disorder that may exist within the twin pair. Secondly, cross-concordance may be the result of an environmental association between the two disorders in the cotwin. Data on comorbid drug and mental disorders in twins with alcoholism will be combined with appropriate control data in order to explore methods of analyzing the etiology of the association between these disorders. Data are available on 169 same-sex twin pairs and analyses will range from simple cross-concordance comparisons to full bivariate cross-twin, cross-trait analysis.